International Journal of Gynecology and Obstetrics 131 (2015) 13–24

Contents lists available at ScienceDirect

International Journal of Gynecology and Obstetrics

j ourna l homepage: www.e lsev ie r .com/ locate / i jgo

FIGO GUIDELINES

FIGO consensus guidelines on intrapartum fetalmonitoring: Cardiotocography☆,★

Diogo Ayres-de-Campos a, Catherine Y. Spong b, Edwin Chandraharan c;for the FIGO Intrapartum Fetal Monitoring Expert Consensus Panel 1

a Medical School, Institute of Biomedical Engineering, S. Joao Hospital, University of Porto, Portugalb Eunice Kennedy Shriver National Institute of Child Health and Human Development, Bethesda, MD, USAc St George’s University Hospitals NHS Foundation Trust, London, UK

1. Introduction

The purpose of this chapter is to assist in the use and interpretationof intrapartum cardiotocography (CTG), as well as in the clinical man-agement of specific CTG patterns. In the preparation of these guidelines,it has been assumed that all necessary resources, both human and ma-terial, required for intrapartum monitoring and clinical managementare readily available. Unexpected complications may occur duringlabor, even in patients without prior evidence of risk, so maternity hos-pitals need to ensure the presence of trained staff, as well as appropriatefacilities and equipment for an expedite delivery (in particular emer-gency cesarean delivery). CTG monitoring should never be regarded asa substitute for good clinical observation and judgement, or as an excusefor leaving the mother unattended during labor.

2. Indications

The evidence for the benefits of continuous CTG monitoring, ascompared with intermittent auscultation, in both low- and high-risklabors is scientifically inconclusive [1,2].When comparedwith intermit-tent auscultation, continuous CTG has been shown to decrease the

☆ Developed by the FIGO Safe Motherhood and Newborn Health Committee;coordinated by Diogo Ayres-de-Campos.★ The views expressed in this document reflect the opinion of the individuals and not

necessarily those of the institutions that they represent.1 Consensus panel: Daniel Surbek (Switzerland⁎), Gabriela Caracostea (Romania⁎), Yves

Jacquemyn (Belgium⁎), Susana Santo (Portugal⁎), Lennart Nordström (Sweden⁎), TuliaTodros (Italy⁎), Branka Yli (Norway⁎), George Farmakidis (Greece⁎), Sandor Valent(Hungary⁎), Bruno Carbonne (France⁎), Kati Ojala (Finland⁎), José Luis Bartha (Spain⁎),Joscha Reinhard (Germany⁎), Anneke Kwee (Netherlands⁎), Romano Byaruhanga(Uganda⁎), Ehigha Enabudoso (Nigeria⁎), John Anthony (South Africa⁎), Fadi Mirza(Lebanon⁎), Tak Yeung Leung (Hong Kong⁎), Ramon Reyles (Philippines⁎), Park In Yang(South Korea⁎), Henry Murray (Australia and New Zealand⁎), Yuen Tannirandorn(Thailand⁎), Krishna Kumar (Malaysia⁎), Taghreed Alhaidary (Iraq⁎), Tomoaki Ikeda(Japan⁎), Ferdusi Begum (Bangladesh⁎), Jorge Carvajal (Chile⁎), José Teppa (Venezuela⁎),Renato Sá (Brazil⁎), Lawrence Devoe (USA⁎⁎), Gerard Visser (Netherlands⁎⁎), RichardPaul (USA⁎⁎), Barry Schifrin (USA⁎⁎), Julian Parer (USA⁎⁎), Philip Steer (UK⁎⁎), VincenzoBerghella (USA⁎⁎), Isis Amer-Wahlin (Sweden⁎⁎), Susanna Timonen (Finland⁎⁎), AustinUgwumadu (UK⁎⁎), João Bernardes (Portugal⁎⁎), Justo Alonso (Uruguay⁎⁎), SabaratnamArulkumaran (UK⁎⁎).⁎ Nominated by FIGO associated national society; ⁎⁎ Invited by FIGO based on literaturesearch.

http://dx.doi.org/10.1016/j.ijgo.2015.06.0200020-7292/© 2015 Published by Elsevier Ireland Ltd. on behalf of International Federation of G

occurrence of neonatal seizures, but no effect has been demonstratedon the incidence of overall perinatalmortality or cerebral palsy. Howev-er, these studies were carried out in the 1970s, 1980s, and early 1990swhere equipment, clinical experience, and interpretation criteria werevery different from current practice, and they were clearly underpow-ered to evaluate differences in major outcomes [3]. These issues arediscussed in more detail in Section 8 of this chapter. In spite of theselimitations, most experts believe that continuous CTG monitoringshould be considered in all situations where there is a high risk of fetalhypoxia/acidosis, whether due to maternal health conditions (such asvaginal hemorrhage and maternal pyrexia), abnormal fetal growthduring pregnancy, epidural analgesia, meconium stained liquor, or thepossibility of excessive uterine activity, as occurs with induced or aug-mented labor. Continuous CTG is also recommended when abnormali-ties are detected during intermittent fetal auscultation. The use ofcontinuous intrapartum CTG in low-risk women is more controversial,although it has become standard of care in many countries. An alterna-tive approach is to provide intermittent CTG monitoring alternatingwith fetal heart rate (FHR) auscultation. There is some evidence to sup-port that this is associated with similar neonatal outcomes in low-riskpregnancies [4]. Intermittent monitoring should be carried out longenough to allow adequate evaluation of the basic CTG features (seebelow). The routine use of admission CTG for low-risk women on en-trance to the labor ward has been associated with an increase in cesar-ean delivery rates and no improvement in perinatal outcomes [5], butstudies were also underpowered to show such differences. In spite ofthe lack of evidence regarding benefit, this procedure has also becomestandard of care in many countries.

3. Tracing acquisition

3.1. Maternal position for CTG acquisition

Maternal supine recumbent position can result in aortocaval com-pression by the pregnant uterus, affecting placental perfusion and fetaloxygenation. Prolonged monitoring in this position should thereforebe avoided. The lateral recumbent, half-sitting, and upright positionsare preferable alternatives [6].

CTG acquisition can be performed by portable sensors that transmitsignals wirelessly to a remote fetal monitor (telemetry). This solution

ynecology and Obstetrics.

14 D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

has the advantage of allowing the mother to move freely during signalacquisition, rather than be restrained to bed or a sofa, and should there-fore be the preferred optionwhen available. Telemetry systems differ inthe maximum distance allowed between patient and monitor for ade-quate signal transmission [7].

3.2. Paper scales for CTG registration and viewing

The horizontal scale for CTG registration and viewing is commonlycalled “paper speed” and available options are usually 1, 2, or 3 cm/min.In many countries throughout the world 1 cm/min is selected, while inthe Netherlands it is usually 2 cm/min, and in North America and Japanit is almost exclusively 3 cm/min. Some experts feel that 1 cm/min pro-vides records of sufficient detail for clinical analysis, and this has the ad-vantage of reducing tracing length. Other experts feel that the smalldetails of CTG tracings are better evaluated using higher papers speeds.The vertical scale used for registration and viewing may also be different,and available alternatives are 20 or 30 bpm/cm.

The paper scales used in each center should be the ones with whichhealthcare professionals are most familiar, because tracing interpreta-tion depends on pattern recognition and these patterns may appearvery different. Inadvertent use of paper scales towhich the staff is unac-customedmay lead to erroneous interpretations of CTG features. For ex-ample, at 3 cm/min variability appears reduced to a clinician familiarwith the 1 cm/min scale, while it may appear exaggerated in the oppo-site situation (see examples below).

3.3. External versus internal FHR monitoring

External FHR monitoring uses a Doppler ultrasound transducer todetect themovement of cardiac structures. The resulting signal requiressignal modulation and autocorrelation to provide adequate quality

Fig 1.Maternal heart rate monitoring in the last 9 minutes of the tracing. External fetal heart ragraph).

recordings [8]. This process results in an approximation of the trueheart rate intervals, but this is considered to be sufficiently accuratefor analysis. External FHRmonitoring is more prone to signal loss, to in-advertent monitoring of the maternal heart rate (Fig. 1) [9], and to sig-nal artefacts such as double-counting (Fig. 2) and half-counting [8],particularly during the second stage of labor. It may also not recordfetal cardiac arrhythmias accurately.

Internal FHR monitoring using a fetal electrode (usually known asscalp electrode, but it can also be applied to the breech) evaluates thetime intervals between successive heart beats by identifying R waveson the fetal electrocardiogram QRS complex, and therefore measuresventricular depolarization cycles. This method provides a more accurateevaluation of intervals between cardiac cycles, but it is more expensivebecause it requires a disposable electrode. It is very important that thefetal electrode is only applied after a clear identification of the presentingpart and that delicate fetal structures such as the sutures and fontanelsare avoided. Internal FHR monitoring requires ruptured membranesand has established contraindications, mainly related to the increasedrisk of vertical transmission of infections. It should not be used in patientswith active genital herpes infection, those who are seropositive to hepa-titis B, C, D, E, or to HIV [10,11], in suspected fetal blood disorders, whenthere is uncertainty about the presenting part, or when artificial ruptureof membranes is inappropriate (i.e. an unengaged presentation). Fetalelectrode placement should also preferably be avoided in very pretermfetuses (under 32 weeks of gestation).

External FHR monitoring is the recommended initial method forroutine intrapartummonitoring, provided that a recording of acceptablequality is obtained, i.e. that the basic CTG features can be identified.Minimum requirements for using this method are that careful reposi-tioning of the probe is carried out during the second stage of labor,that in all atypical FHR tracings maternal heart rate monitoring isruled out (see below), and if any doubt remains, fetal auscultation,

te monitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom

Fig 2. Double-counting of the fetal heart rate during decelerations (arrows). External fetal heart rate monitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min(bottom graph).

15D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

ultrasound evaluation, or internal FHR monitoring are performed. If anacceptable record cannot be obtained with external monitoring or if acardiac arrhythmia is suspected, then internal monitoring should beused, in the absence of the previously mentioned contraindications.

3.4. External versus internal monitoring of uterine contractions

Externalmonitoring of uterine contractions using a tocodynamometer(toco) evaluates increased myometrial tension measured through theabdominal wall. Incorrect placement, reduced tension applied to thesupporting elastic band, or abdominal adiposity may result in failed orinadequate registration of contractions. In addition, this technologyonly provides accurate information on the frequency of contractions.It is not possible to extract reliable information regarding the intensityand duration of contractions, nor on basal uterine tone.

Internal monitoring of uterine contractions using an intrauterinecatheter provides quantitative information on the intensity and durationof contractions, as well as on basal uterine tone, but it is more expensiveas the catheter is disposable, and requires ruptured membranes.Contraindications include uterine hemorrhage of unknown cause andplacenta previa. It may also be associatedwith a small risk of fetal injury,placental hemorrhage, uterine perforation, and infection [12]. Theroutine use of intrauterine pressure catheters has not been shown tobe associated with improved outcomes in induced and augmentedlabor [13], and so it is not recommended for routine clinical use.

3.5. Simultaneous monitoring of the maternal heart rate

Simultaneous monitoring of the maternal heart rate (MHR) can beuseful in specific maternal health conditions and in caseswhere it is dif-ficult to distinguish betweenmaternal and fetal heart rates (for examplecomplete fetal heart block) [9]. Some CTG monitors provide the

possibility of continuous MHR monitoring, either by electrocardiogra-phy or pulse oximetry. In some recent models, the latter technologyhas been incorporated in the tocodynamometer, allowing continuousMHR monitoring without the use of additional equipment. Providingthat the technology is available and does not cause discomfort to themother, simultaneous MHR monitoring should be considered whenperforming continuous CTG, especially during the second stage oflabor, when tracings show accelerations coinciding with contractionsand expulsive efforts [9], or when the MHR is elevated.

3.6. Monitoring of twins

Continuous external FHR monitoring of twin gestations duringlabor should preferably be performed with dual channel monitors thatallow simultaneous monitoring of both FHRs, as duplicate monitoringof the same twin may occur and this can be picked up by observingalmost identical tracings. Somemonitors have embedded algorithmsto alarm when this situation is suspected. During the second stage oflabor, external FHR monitoring of twins is particularly affected bysignal loss, and for this reason some experts believe that the present-ing twin should preferably be monitored internally for better signalquality [14], if no contraindications to fetal electrode placement arepresent. Other experts believe that external monitoring of bothtwins is acceptable, provided that distinct and good quality FHR signalscan be obtained.

3.7. Storage of tracings

All CTG tracings need to be identifiedwith the patient name, place ofrecording, “paper speed,” and date and time when acquisition startedand ended. In hospitals where paper CTG recordings are used, the lattershould be considered as part of the patient record and preserved as

16 D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

such. In hospitals using digital CTG archives [15], a secure file backupsystem needs to be in place, and all tracings should be readily availablefor review by the clinical staff.

4. Analysis of tracings

CTG analysis starts with the evaluation of basic CTG features(baseline, variability, accelerations, decelerations, and contractions)followed by overall CTG classification.

4.1. Evaluation of basic CTG features

4.1.1. BaselineThis is themean level of themost horizontal and less oscillatory FHR

segments. It is estimated in time periods of 10minutes and expressed inbeats per minute (bpm). The baseline value may vary between subse-quent 10-minute sections.

In tracings with unstable FHR signals, review of previous segmentsand/or evaluation of longer time periods may be necessary to estimatethe baseline [16], in particular during the second stage of labor and toidentify the fetal behavioral state of active wakefulness (Fig. 3) thatcan lead to erroneously high baseline estimation.

Normal baseline: a value between 110 and 160 bpm.Preterm fetuses tend to have values toward the upper end of thisrange and post-term fetuses towards the lower end. Some expertsconsider the normal baseline values at term to be between 110-150 bpm.Tachycardia: a baseline value above 160 bpm lasting more than10 minutes.Maternal pyrexia is themost frequent cause of fetal tachycardia, and itmay be of extrauterine origin or associatedwith intrauterine infection.Epidural analgesia may also cause a rise in maternal temperature

Fig 3. Fetal behavioral state of active wakefulness. This patternmay lead to erroneously high baitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom graph).

resulting in fetal tachycardia [17]. In the initial stages of a non-acute fetal hypoxemia, catecholamine secretion may also result intachycardia. Other less frequent causes are the administration ofbeta-agonist drugs [18] (salbutamol, terbutaline, ritodrine, fenoterol),parasympathetic blockers (atropine, scopolamine), and fetal arrhyth-mias such as supraventricular tachycardia and atrial flutter.Bradycardia: a baseline value below 110 bpm lasting more than10 minutes.Values between 100 and 110 bpm may occur in normal fetuses,especially in postdate pregnancies. Maternal hypothermia [19],administration of beta-blockers [20], and fetal arrhythmias such asatrioventricular block are other possible causes.

4.1.2. VariabilityThis refers to the oscillations in the FHR signal, evaluated as

the average bandwidth amplitude of the signal in 1-minute segments.

Normal variability: a bandwidth amplitude of 5−25 bpm.Reduced variability: a bandwidth amplitude below 5 bpm for morethan 50 minutes in baseline segments [21] (Figs. 4 and 5), or formore than 3 minutes during decelerations [22] (see Figs. 8 and 9).Reduced variability can occur due to central nervous systemhypoxia/acidosis and resulting decreased sympathetic and parasym-pathetic activity, but it can also be due to previous cerebral injury[23], infection, administration of central nervous systemdepressantsor parasympathetic blockers. During deep sleep, variability is usuallyin the lower range of normality, but the bandwidth amplitude is sel-dom under 5 bpm. There is a high degree of subjectivity in the visualevaluation of this parameter, and therefore careful re-evaluation isrecommended in borderline situations. Following an initially normalCTG, reduced variability due to hypoxia is very unlikely to occur dur-ing labor without preceding or concomitant decelerations and a risein the baseline.

seline estimation if it is identified at the top of accelerations. External fetal heart rate mon-

Fig 4. Reduced variability. External fetal heart rate monitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom graph).

Fig 5. Reduced variability. The baseline is affected by contractions causing decreases in fetal heart rate that are close to fulfilling the criteria for decelerations, but the bandwidth remainsreduced. Internal fetal heart rate monitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom graph).

17D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

18 D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

Increased variability (saltatory pattern): a bandwidth value ex-ceeding 25 bpm lasting more than 30 minutes (Fig. 6).The pathophysiology of this pattern is incompletely understood, butit may be seen linked with recurrent decelerations, when hypoxia/acidosis evolves very rapidly. It is presumed to be caused by fetalautonomic instability/hyperactive autonomic system [24].

4.1.3. AccelerationsAbrupt (onset to peak in less than 30 seconds) increases in FHR

above the baseline, of more than 15 bpm in amplitude, and lastingmore than 15 seconds but less than 10 minutes.

Most accelerations coincidewith fetalmovements and are a sign of aneurologically responsive fetus that does not have hypoxia/acidosis.Before 32 weeks of gestation, their amplitude and frequency may belower (10 seconds and 10 bpm of amplitude). After 32−34 weeks,with the establishment of fetal behavioral states, accelerationsrarely occur during periods of deep sleep, which can last up to50 minutes [21]. The absence of accelerations in an otherwise normalintrapartumCTG is of uncertain significance, but it is unlikely to indicatehypoxia/acidosis. Accelerations coinciding with uterine contractions,especially in the second stage of labor, suggest possible erroneous re-cording of thematernal heart rate, since the FHRmore frequently decel-erates with a contraction, while the maternal heart rate typicallyincreases [9].

4.1.4. DecelerationsDecreases in the FHR below the baseline, of more than 15 bpm in

amplitude, and lasting more than 15 seconds.

Early decelerations: decelerations that are shallow, short-lasting,with normal variability within the deceleration and are coincidentwith contractions. They are believed to be caused by fetal head com-pression [25] and do not indicate fetal hypoxia/acidosis.

Fig 6. Increased variability: saltatory pattern. Internal fetal heart rate monitoring at 1 cm

Variable decelerations (V-shaped): decelerations that exhibit arapid drop (onset to nadir in less than 30 seconds), good variabilitywithin the deceleration, rapid recovery to the baseline, varyingsize, shape, and relationship to uterine contractions (Fig. 7).Variable decelerations constitute the majority of decelerations dur-ing labor, and they translate a baroreceptor-mediated response toincreased arterial pressure, as occurs with umbilical cord compres-sion [26]. They are seldom associated with an important degree offetal hypoxia/acidosis, unless they evolve to exhibit a U-shaped com-ponent, reduced variabilitywithin the deceleration (see late deceler-ations below), and/or their individual duration exceeds 3 minutes[22,27] (see prolonged decelerations below).Late decelerations (U-shaped and/or with reduced variability):decelerations with a gradual onset and/or a gradual return tothe baseline and/or reduced variability within the deceleration(Fig. 8). Gradual onset and return occurs when more than30 seconds elapses between the beginning/end of a decelerationand its nadir. When contractions are adequately monitored, late de-celerations start more than 20 seconds after the onset of a contrac-tion, have a nadir after the acme, and a return to the baseline afterthe end of the contraction.These decelerations are indicative of a chemoreceptor-mediated re-sponse to fetal hypoxemia [25,27]. In the presence of a tracing withno accelerations and reduced variability, the definition of late decel-erations also includes those with an amplitude of 10−15 bpm.Prolonged decelerations: lasting more than 3 minutes.These are likely to include a chemoreceptor-mediated component andthus to indicate hypoxemia. Decelerations exceeding 5 minutes, withFHR maintained at less than 80 bpm and reduced variability withinthe deceleration (Fig. 9), are frequently associated with acute fetalhypoxia/acidosis [22,28–30] and require emergent intervention.

/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom graph).

Fig 7. Variable decelerations. Internal fetal heart rate monitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom graph).

Fig 8. Late decelerations in the second half of the tracing. External fetal heart rate monitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom graph).

19D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

Fig 9. Prolonged deceleration. External fetal heart rate monitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom graph).

20 D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

4.1.5. Sinusoidal patternA regular, smooth, undulating signal, resembling a sine wave, with

an amplitude of 5−15 bpm, and a frequency of 3−5 cycles per minute.This pattern lasts more than 30 minutes, and coincides with absentaccelerations (Fig. 10).

The pathophysiological basis of the sinusoidal pattern is incompletelyunderstood, but it occurs in association with severe fetal anemia, as isfound in anti-D alloimmunization, fetal-maternal hemorrhage, twin-to-twin transfusion syndrome, and ruptured vasa previa. It has also been de-scribed in cases of acute fetal hypoxia, infection, cardiac malformations,hydrocephalus, and gastroschisis [31].

4.1.6. Pseudosinusoidal patternA pattern resembling the sinusoidal pattern, but with a more jagged

“saw-tooth” appearance, rather than the smooth sine-wave form(Fig. 11). Its duration seldom exceeds 30minutes and it is characterizedby normal patterns before and after.

This pattern has been described after analgesic administration to themother, and during periods of fetal sucking and othermouthmovements[32]. It is sometimes difficult to distinguish the pseudosinusoidal patternfrom the true sinusoidal pattern, leaving the short duration of the formeras the most important variable to discriminate between the two.

4.1.7. Fetal behavioral statesThis refers to periods of fetal quiescence reflecting deep sleep

(no eye movements), alternating with periods of active sleep (rapideye movements) and wakefulness [33,34]. The occurrence of differentbehavioral states is a hallmark of fetal neurological responsivenessand absence of hypoxia/acidosis. Deep sleep can last up to 50 minutes[21] and is associated with a stable baseline, very rare accelerations,and borderline variability. Active sleep is the most frequent behavioralstate, and is represented by a moderate number of accelerations and

normal variability. Active wakefulness is rarer and represented by alarge number of accelerations and normal variability (Fig. 1). In the lat-ter pattern, accelerations may be so frequent as to cause difficulties inbaseline estimation (see Fig. 1). Transitions between the different pat-terns become clearer after 32−34 weeks of gestation, consequent tofetal nervous system maturation.

4.1.8. ContractionsThese are bell-shaped gradual increases in the uterine activity signal

followed by roughly symmetric decreases, with 45−120 seconds intotal duration.

Contractions are essential for the progression of labor, but they com-press the vessels running inside the myometrium and may transientlydecrease placental perfusion and/or cause umbilical cord compression[35]. With the tocodynamometer, only the frequency of contractionscan be reliably evaluated, but increased intensity and duration canalso contribute to FHR changes.

Tachysystole: This represents an excessive frequency of contrac-tions and is defined as the occurrence of more than five contractionsin 10 minutes, in two successive 10-minute periods, or averagedover a 30-minute period.

5. Tracing classification

Tracing classification requires a previous evaluation of basic CTG fea-tures (see above). Tracings should be classified into one of three classes:normal, suspicious, or pathological, according to the criteria presentedin Table 1. Other classification systems including a larger number oftiers are recommended by some experts [36,37]. Due to the changingnature of CTG signals during labor, re-evaluation of the tracing shouldbe carried out at least every 30 minutes.

Fig 10. Sinusoidal pattern. External fetal heart rate monitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom graph).

Fig 11. Pseudosinusoidal pattern. External fetal heart rate monitoring at 1 cm/min (top graph), 2 cm/min (middle graph), and 3 cm/min (bottom graph).

21D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

Table 1Cardiotocography classification criteria, interpretation, and recommended management.a

Normal Suspicious Pathological

Baseline 110−160 bpm Lacking at least one characteristic of normality, butwith no pathological features

b100 bpm

Variability 5−25 bpm Lacking at least one characteristic of normality, butwith no pathological features

Reduced variability, increased variability, or sinusoidalpattern

Decelerations No repetitiveb decelerations Lacking at least one characteristic of normality, butwith no pathological features

Repetitiveb late or prolonged decelerations duringN30 min or 20 min if reduced variability, or oneprolonged deceleration with N5 min

Interpretation Fetus with no hypoxia/acidosis Fetus with a low probability of havinghypoxia/acidosis

Fetus with a high probability of having hypoxia/acidosis

Clinical management No intervention necessary to improvefetal oxygenation state

Action to correct reversible causes if identified,close monitoring or additional methods to evaluatefetal oxygenation [49]

Immediate action to correct reversible causes,additional methods to evaluate fetal oxygenation[49], or if this is not possible expedite delivery. Inacute situations (cord prolapse, uterine rupture, orplacental abruption) immediate delivery should beaccomplished.

a The presence of accelerations denotes a fetus that does not have hypoxia/acidosis, but their absence during labor is of uncertain significance.b Decelerations are repetitive in nature when they are associated with more than 50% of uterine contractions [29].

22 D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

6. Clinical decision

Several factors, including gestational age and medication adminis-tered to themother, can affect FHR features (see above), so CTG analysisneeds to be integrated with other clinical information for a comprehen-sive interpretation and adequate management. As a general rule, if thefetus continues tomaintain a stable baseline and a reassuring variability,the risk of hypoxia to the central organs is very unlikely. However, thegeneral principles that should guide clinical management are outlinedin Table 1.

7. Action in situations of suspected fetal hypoxia/acidosis

When fetal hypoxia/acidosis is anticipated or suspected (suspiciousand pathological tracings), and action is required to avoid adverse neo-natal outcome, this does not necessarily mean an immediate cesareandelivery or instrumental vaginal delivery. The underlying cause for theappearance of the pattern can frequently be identified and the situationreversed, with subsequent recovery of adequate fetal oxygenation andthe return to a normal tracing.

Excessive uterine activity is themost frequent cause of fetal hypoxia/acidosis [35] and it can be detected by documenting tachysystole in theCTG tracing and/or palpating the uterine fundus. It can usually be re-versed by reducing or stopping oxytocin infusion, removing adminis-tered prostaglandins if possible, and/or starting acute tocolysis withbeta-adrenergic agonists (salbutamol, terbutaline, ritodrine) [38–40],atosiban [41], or nitroglycerine [42]. During the second stage of labor,maternal pushing efforts can also contribute to fetal hypoxia/acidosisand the mother can be asked to stop pushing until the situationis reversed.

Aortocaval compression can occur in the supine position and lead toreduced placental perfusion. Excessive uterine activitymay also be asso-ciated with the supine position [43,44], possibly due to the stimulationof the sacral plexus by the uterine weight. In these cases, turning themother to her side is frequently followed by normalization of the CTGpattern. Transient cord compression is another common cause of CTGchanges (variable decelerations), and these can sometimes be revertedby changing thematernal position or by performing amnioinfusion [45].

Sudden maternal hypotension can also occur during labor, usuallyafter epidural or spinal analgesia [46], and it is usually reversible byrapid fluid administration and/or an intravenous ephedrine bolus.Other less frequent complications affecting the maternal respiration,maternal circulation, placenta, umbilical cord, or the fetal circulationcan also result in fetal hypoxia/acidosis [35], and their management isbeyond the scope of this document.

Oxygen administration to themother is widely used with the objec-tive of improving fetal oxygenation and consequently normalizing CTG

patterns, but there is no evidence from randomized clinical trials thatthis intervention, when performed in isolation, is effectivewhenmater-nal oxygenation is adequate [47]. Intravenous fluids are also commonlyused for the purpose of improving CTG patterns, but again there is noevidence from randomized clinical trials to suggest that this interven-tion is effective in normotensive women [48].

Good clinical judgement is required to diagnose the underlyingcause for a suspicious or pathological CTG, to judge the reversibility ofthe conditions with which it is associated, and to determine the timingof delivery, with the objective of avoiding prolonged fetal hypoxia/acidosis, as well as unnecessary obstetric intervention. Additionalmethodsmay be used to evaluate fetal oxygenation [49]. When a suspi-cious or worsening CTG pattern is identified, the underlying causeshould be addressed before a pathological tracing develops. If the situa-tion does not revert and the pattern continues to deteriorate, consider-ation needs to be given for further evaluation or rapid delivery if apathological pattern ensues.

During the second stage of labor, due to the additional effect of ma-ternal pushing, hypoxia/acidosis may develop more rapidly. Therefore,urgent action should be undertaken to relieve the situation, includingdiscontinuation of maternal pushing, and if there is no improvement,delivery should be expedited.

8. Limitations of cardiotocography

Cardiotocography has well-documented limitations, and it is neces-sary to be aware of these for safe use of the technology.

It has beenwell demonstrated that CTG analysis is subject to consid-erable intra- and interobserver disagreement, even when experiencedclinicians use widely accepted guidelines [50–52]. The main aspectsthat are prone to observer disagreement are the identification and clas-sification of decelerations, the evaluation of variability [51], and theclassification of tracings as suspicious and pathological [51,52]. The sub-jectivity of observer analysis has also been demonstrated in retrospec-tive audit of tracings, where CTG features are frequently assessed to bemore abnormal in cases with known adverse neonatal outcome [53].

Many studies have evaluated the ability of suspicious and patholog-ical CTGs to predict the occurrence of hypoxia/acidosis. Different CTG in-terpretation criteria, different intervals between tracing abnormalityand birth, and different criteria to define adverse outcome have beenused, resulting in mixed findings [54]. However, it is recognized thathypoxia/acidosis has not been documented shortly after a normal CTGtracing. On the other hand, suspicious and pathological tracings have alimited capacity to predict metabolic acidosis and low Apgar scores,i.e. a large percentage of cases with suspicious and pathological tracingsdo not have these outcomes [54].While there is a strong association be-tween certain FHR patterns and hypoxia/acidosis, their capacity to

23D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

discriminate between newborns with or without metabolic acidosis islimited. Thus, they are sensitive indicators, but have a low specificityand low positive predictive value. However, it should not be forgottenthat the aim of intrapartum fetal monitoring is to identify situationsthat precede hypoxia/acidemia so as to avoid fetal injury. The subjectiv-ity of CTG interpretation and the fact that hypoxia is a continuum thatmay not reach the threshold ofmetabolic acidosis or injury are probablyimportant contributing factors to these limitations.

A large number of randomized controlled trials have been conductedcomparing continuous CTG monitoring with intermittent auscultationas screening methods for fetal hypoxia/acidosis during labor, in bothlow- and high-risk women [1,2]. However, these trials took place inthe 1970s, 1980s, and early 1990s, and used different CTG interpretationcriteria, so it is difficult to establish how their results relate to currentclinical practice. With these limitations in mind, they indicate a limitedbenefit of continuous CTG for fetal monitoring in all women duringlabor, as the only significant improvement was a 50% reduction in neo-natal seizures (hypoxic-ischemic encephalopathy was not evaluated inmost trials), and no differences were found in the incidences of overallperinatal mortality and cerebral palsy. However, it is widely recognizedthat the trials were underpowered to detect differences in these out-comes [3]. Only a small proportion of perinatal deaths and cerebralpalsies are caused by intrapartum hypoxia/acidosis, so a large numberof cases are needed to show any benefit. On the other hand, continuousCTG was associated with a 63% increase in cesarean delivery and a 15%increase in instrumental vaginal deliveries [1].

Unnecessary obstetric intervention confers additional risks for themother and newborn [55,56], and the former may result from poorCTG interpretation, limited knowledge of the pathophysiology of fetaloxygenation, and inadequate clinical management. It is recognizedthat, for consistent implementation, clinical guidelines need to beas simple and objective as possible, to allow rapid decision-makingeven in complex and stressful situations. In addition, regular and struc-tured training of the labor ward staff is essential to ensure proper use ofthis technology.

Conflict of interest

The authors have no conflicts of interest to declare.

References

[1] Alfirevic Z, Devane D, Gyte GM. Continuous cardiotocography (CTG) as a form ofelectronic fetal monitoring (EFM) for fetal assessment during labour. CochraneDatabase Syst Rev 2013;5:CD006066.

[2] Vintzileos AM, Nochimson DJ, Guzman ER, Knuppel RA, Lake M, Schifrin BS. Inter-mittent electronic fetal heart rate monitoring versus intermittent auscultation: ameta-analysis. Obstet Gynecol 1995;85(1):149–55.

[3] Royal College of Obstetricians and Gynaecologists. The use of electronic fetal moni-toring. Evidence-based clinical guideline, number 8. London: RCOG Press; 2001.

[4] Herbst A, Ingemarsson I. Intermittent versus continuous electronic fetal monitoringin labour: a randomized study. Br J Obstet Gynaecol 1994;101(8):663–8.

[5] Devane D, Lalor JG, Daly S, McGuire W, Smith V. Cardiotocography versus intermit-tent auscultation of fetal heart on admission to labour ward for assessment of fetalwellbeing. Cochrane Database Syst Rev 2012;2:CD005122.

[6] Carbonne B, Benachi A, Leveque ML, Cabrol D, Papiernik E. Maternal position duringlabor: effects on fetal oxygen saturation measured by pulse oximetry. ObstetGynecol 1996;88(5):797–800.

[7] Di Renzo GC, Luzi G, Caserta G, Clerici G, Cosmi EV. The role of telemetry in perinatalmonitoring. J Perinat Med 1994;22(6):517–22.

[8] Carter MC. Signal processing and display–cardiotocographs. Br J Obstet Gynaecol1993;100(Suppl. 9):21–3.

[9] Nurani R, Chandraharan E, Lowe V, Ugwumadu A, Arulkumaran S. Misidentificationof maternal heart rate as fetal on cardiotocography during the second stage oflabour: the role of the fetal electrocardiograph. Acta Obstet Gynecol Scand 2012;91(12):1428–32.

[10] Maiques V, Garcia-Tejedor A, Perales A, Navarro C. Intrapartum fetal invasive proce-dures and perinatal transmission of HIV. Eur J Obstet Gynecol Reprod Biol 1999;87(1):63–7.

[11] Kaye EM, Dooling EC. Neonatal herpes simplexmeningoencephalitis associatedwithfetal monitor scalp electrodes. Neurology 1981;31(8):1045–7.

[12] Handwerker SM, Selick AM. Placental abruption after insertion of catheter tip intra-uterine pressure transducers: a report of four cases. J ReprodMed 1995;40(12):845–9.

[13] Bakker JJ, Janssen PF, van Halem K, van der Goes BY, Papatsonis DN, van des Post JA,et al. Internal versus external tocodynamometry during induced or augmentedlabour. Cochrane Database Syst Rev 2013;8:CD006947.

[14] Bakker PC, Colenbrander GJ, Verstraeten AA, Van Geijn HP. Quality of intrapartumcardiotocography in twin deliveries. Am J Obstet Gynecol 2004;191(6):2114–9.

[15] Nunes I, Ayres-de-Campos D, Figueiredo C, Bernardes J. An overview of central fetalmonitoring systems in labour. J Perinat Med 2013;41(1):93–9.

[16] Ayres-de-Campos D, Bernardes J, Marsal K, Nickelsen C, Makarainen L, Banfield P,et al. Can the reproducibility of fetal heart rate baseline estimation be improved?Eur J Obstet Gynecol Reprod Biol 2004;112(1):49–54.

[17] Segal S. Labor epidural analgesia and maternal fever. Anesth Analg 2010;111(6):1467–75.

[18] Neilson JP, West HM, Dowswell T. Betamimetics for inhibiting preterm labour.Cochrane Database Syst Rev 2014;2:CD004352.

[19] Jadhon ME, Main EK. Fetal bradycardia associated with maternal hypothermia.Obstet Gynecol 1988;72(3 Pt 2):496–7.

[20] Boutroy MJ. Fetal and neonatal effects of the beta-adrenoreceptor blocking agents.Dev Pharmacol Ther 1987;10(3):224–31.

[21] Suwanrath C, Suntharasaj T. Sleep–wake cycles in normal fetuses. Arch GynecolObstet 2010;281(3):449–54.

[22] Hamilton E, Warrick P, O’Keeffe D. Variable decelerations: do size and shapematter?J Matern Fetal Neonatal Med 2012;25(6):648–53.

[23] Nelson KB, Dambrosia JM, Ting TY, Grether JK. Uncertain value of electronic fetalmonitoring in predicting cerebral palsy. N Engl J Med 1996;334(10):613–8.

[24] Nunes I, Ayres-de-Campos D, Kwee A, Rosen KG. Prolonged saltatory fetal heart ratepattern leading to newbornmetabolic acidosis. Clin Exp Obstet Gynecol 2014;41(5):507–11.

[25] Court DJ, Parer JT. Experimental studies of fetal asphyxia and fetal heart rate inter-pretation. In: Nathanielsz PW, Parer JT, editors. Research in Perinatal Medicine (I).New York: Perinatalogy Press; 1984. p. 113–69.

[26] Ball RH, Parer JT. The physiologic mechanisms of variable decelerations. Am J ObstetGynecol 1992;166(6 Pt 1):1683–9.

[27] Holzmann M, Wretler S, Cnattingius S, Nordstrom L. Cardiotocography patterns andrisk of intrapartum fetal acidemia. J Perinat Med 2015;43(4):473–9.

[28] Westgate JA,Wibbens B, Bennet L,Wassink G, Parer JT, Gunn AJ. The intrapartumde-celeration in center stage: a physiologic approach to the interpretation of fetal heartrate changes in labor. Am J Obstet Gynecol 2007;197(3):236.e1–236.e11.

[29] Cahill AG, Roehl KA, Odibo AO, Macones GA. Association and prediction of neonatalacidemia. Am J Obstet Gynecol 2012;207(3):206.e1–8.

[30] Takano Y, Furukawa S, Ohashi M, Michikata K, Sameshima H, Ikenoue T. Fetal heartrate patterns related to neonatal brain damage and neonatal death in placentalabruption. J Obstet Gynaecol Res 2013;39(1):61–6.

[31] Modanlou HD, Murata Y. Sinusoidal fetal heart rate pattern: reappraisal of its defini-tion and clinical significance. J Obstet Gynaecol Res 2004;30(3):169–80.

[32] Graça LM, Cardoso CG, Calhaz-Jorge C. An approach to interpretation and classifica-tion of sinusoidal fetal heart rate patterns. Eur J Obstet Gynecol Reprod Biol 1988;27(3):203–12.

[33] Nijhuis JG, Prechtl HF, Martin CB, Bots RS. Are there behavioural states in the humanfetus? Early Hum Dev 1982;6(2):177–95.

[34] de Vries JI, Visser GH, Prechtl HF. The emergence of fetal behaviour II. Quantitativeaspects. Early Hum Dev 1985;12(2):99–120.

[35] Ayres-de-Campos D, Arulkumaran S, FIGO Intrapartum Fetal Monitoring ExpertConsensus Panel. FIGO consensus guidelines on intrapartum fetal monitoring:Physiology of fetal oxygenation and the main goals of intrapartum fetal monitoring.Int J Gynecol Obstet 2015;131:5–8 [in this issue].

[36] Parer JT, Ikeda T. A framework for standardized management of intrapartum fetalheart rate patterns. Am J Obstet Gynecol 2007;197(1):26.e1–6.

[37] Coletta J, Murphy E, Rubeo Z, Gyamfi-Bannerman C. The 5-tier system of assessingfetal heart rate tracings is superior to the 3-tier system in identifying fetal acidemia.Am J Obstet Gynecol 2012;206(3):226.e1–5.

[38] Heuser CC, Knight S, Esplin S, Eller AG, Holmgren CM, Richards D, et al. Tachysystolein term labor: incidence, risk factors, outcomes, and effect on fetal heart tracings. AmJ Obstet Gynecol 2013;209(1):32.e1–6.

[39] Briozzo L, Martinez A, Nozar M, Fiol V, Pons J, Alonso J. Tocolysis and delayed deliv-ery versus emergency delivery in cases of non-reassuring fetal status during labor. JObstet Gynaecol Res 2007;33(3):266–73.

[40] Heus R, Mulder EJ, Derks JB, Visser GH. Acute tocolysis for uterine activity reductionin term labor: a review. Obstet Gynecol Surv 2008;63(6):383–8.

[41] Heus R, Mulder EJ, Derks JB, Kurver PH, van Wolfswinkel L, Visser GH. A prospectiverandomized trial of acute tocolysis in term labour with atosiban or ritodrine. Eur JObstet Gynecol Reprod Biol 2008;139(2):139–45.

[42] Pullen KM, Riley ET, Waller SA, Taylor L, Caughey AB, Druzin ML, et al. Randomizedcomparison of intravenous terbutaline vs nitroglycerin for acute intrapartum fetalresuscitation. Am J Obstet Gynecol 2007;197(4):414.e1–6.

[43] Williams EA. Abnormal uterine action during labour. J Obstet Gynaecol Br Emp1952;59:635–41.

[44] Caldeyro-Barcia R, Noriega-Guerra L, Cibils LA, Alvarez H, Poseiro JJ, Pose ST. Effect ofposition changes on the intensity and frequency of uterine contractions duringlabor. Am J Obstet Gynecol 1960;80:284–90.

[45] Hofmeyr GJ, Lawrie TA. Amnioinfusion for potential or suspected umbilical cordcompression in labour. Cochrane Database Syst Rev 2012;1:CD000013.

[46] Simmons SW, Taghizadeh N, Dennis AT, Hughes D, Cyna AM. Combined spinal-epidural versus epidural analgesia in labour. Cochrane Database Syst Rev 2012;10:CD003401.

[47] Fawole G, Hofmeyr GJ. Maternal oxygen administration for fetal distress. CochraneDatabase Syst Rev 2012;12:CD000136.

24 D. Ayres-de-Campos et al. / International Journal of Gynecology and Obstetrics 131 (2015) 13–24

[48] Simpson KR, James DC. Efficacy of intrauterine resuscitation techniques in improvingfetal oxygen status during labor. Obstet Gynecol 2005;105(6):1362–8.

[49] Visser GH, Ayres-de-Campos D, FIGO Intrapartum Fetal Monitoring ExpertConsensus Panel. FIGO consensus guidelines on intrapartum fetal monitoring:Adjunctive technologies. Int J Gynecol Obstet 2015;131:25–9 [in this issue].

[50] Paneth N, Bommarito M, Stricker J. Electronic fetal monitoring and later outcome.Clin Invest Med 1993;16(2):159–65.

[51] Ayres-de-Campos D, Bernardes J, Costa-Pereira A, Pereira-Leite L. Inconsistencies inclassification by experts of cardiotocograms and subsequent clinical decision. BJOG1999;106(12):1307–10.

[52] Blackwell SC, Grobman WA, Antoniewicz L, Hutchinson M, Gyamfi-Bannerman C.Interobserver and intraobserver reliability of the NICH 3-tier fetal heart rate inter-pretation system. Am J Obstet Gynecol 2011;205(4):378.e1–5.

[53] Ayres-de-Campos D, Arteiro D, Costa-Santos C, Bernardes J. Knowledge of adverse neo-natal outcome alters clinicians' interpretation of the intrapartum cardiotocograph. BJOG2011;118(8):978–84.

[54] Spencer JA. Clinical overview of cardiotocography. Br J Obstet Gynaecol 1993;100(Suppl. 9):4–7.

[55] National Institute for Health and Clinical Excellence. Caesarean section. NICE clinicalguideline 13. April 2004 http://www.nice.org.uk/guidance/cg13/resources/guid-ance-caesarean-section-pdf. Accessed August 1, 2014.

[56] O’Mahony F, HofmeyrGJ,MenonV. Choice of instruments for assisted vaginal delivery.Cochrane Database Syst Rev 2010;11:CD005455.